Gene therapy is expected to become a common medical practice in years to come. Several vector systems are being developed to affect gene transfer to correct inherited or acquired diseases, such as cancer and AIDS. Lentiviral vectors are specially well suited for gene transfer into non-dividing as well as dividing cells, including stem cells. Exploiting our expertise in human lentivirus (HIV-1 and HIV-2) gene regulation, we have designed novel HIV-1 and HIV-2 vectors carrying 'real' genes and are evaluating their efficacy of gene transfer in a variety of cells. These 'real' genes represent different disease entitities, target cell specificities, and animal model availabilities. They include the Bax gene for apoptosis in neuroblastoma, the viral Vpr gene (cell cycle arrest) and caspase-3 gene (apoptosis) for one-two punch in neoplasia, the AADC gene in Parkinson's disease for targeting a biochemical pathway defect, and the alpha-galactosidase (AGA) gene in Fabry disease for correcting an in-born error of metabolism. The vectors that could be produced in reasonable titer demonstrably affected gene transfer in appropriate target cells; AADC imparting the capacity to convert L-dopa into dopamine in primary fetal brain cells and AGA clearing glycolipid deposits in fibroblasts from Fabry patients. Generation of Bax vectors present an interesting dilemma. The objective was to create a vector that was strongly apoptotic, but such a vector turned out to be deleterious to the producer cells. This is an example of general problem where a transgene affects the components of the packaging machinery or those of the vector itself. We are attempting to find a general solution to this problem. One strategy is to utilize the viral LTR promoter for viral RNA expression and a tissue-selective internal promoter for transgene expression, such as the human telomerase promoter for targeting cancer cells. Looking to the future and to exploit the full translational potential of lentiviral vectors, two issues are of paramount importance: regulation and safety. To acquire the ability to regulate in vivo vector expression from outside, we are screening in medium and high throughput formats chemical libraries to search for drug-like molecules. Both the viral LTR and human telomerase promoter are the targets for this molecular targets research. As a proof of principal, the initial search is for a small molecule homolog of the viral Tat, which normally regulates viral LTR expression. Because we are the original discoverers of the unusual structure and function of the tat gene, this exercise is appropriate and exciting. It is made more exciting as the search is a collaborative effort of divisions cutting across intramural and extramural boundaries. For safety, we are continuing over efforts to create chimeric HIV-1 and HIV-2 vectors. Our recent finding of the non-reciprocal cross packaging is reminiscent of our previous observations on Tat mediated one-way cross-transactivation between HIV-1 and HIV-2. We believe both phenomenon have similar basis.